Cloud link (Demo): https://inuwamobarak-pavement-degradation-resnet50-home-0ec0l3.streamlit.app/
According to Asphalt Magazine, before the appropriate repair strategy can be applied to a distressed asphalt pavement, the type and extent of the deterioration must be understood, and the cause of the distress must be identified. This will help to know how to appropriately implement the best repiar strategy.
Created my dataset by Web Scraping over 10,000 images of pavement images for the project.
Using PyTorch and TensorFlow, I built a model for classifying 3 prevalent asphalt degredations including Fatigue Cracking, Linear Cracking and Potholes. This provides engineers a tool to quickly know how to efficiently foster the approate solution.
The deployed model is built using PyTorch(https://pytorch.org/), an open source machine learning framework for building state of the art models to production deployment.
Transfer learning was used to transfer pretrained weights from the resnt50 architecture after webscraping about 10,000 images.
My Dataset link: https://drive.google.com/drive/folders/1HbsTu1BuMpTGQpnS6lfQODgEfvj7n-Fo?usp=share_link
Contact me for other personal powerful datasets.